Alfa receiver Tsys using the old if/lo system.
22apr04
Alfa was installed on 21apr04. We (ganesh , lisa, and
myself) measured the alfa system temperature using the alfa high cal. Since
the alfa iflo box had not been installed yet, we used the lband wide cables
to bring 1 pixel at a time down to the control room using the regular iflo
system. Blank sky was tracked during the measurements (ra=11:00:00, dec=23:00:00).
The za stayed below 12 degrees za during the measurements. There was no
rf filter used and the IF was 750 Mhz and a high side LO.
The calon/offs were taken (3 seconds on, 3 seconds
off). The interim correlator was setup to take 4*25 Mhz with 1024
channels per band giving a frequency resolution of 48 Khz after hanning
smoothing. The band 1220 to 1520 was covered in 3 100 Mhz sets. The entire
band (1220 to 1520) was repeated 3 times. The IF bandpass was removed by
dividing the cal off spectrum by the cal deflection TsysCalOff/(TsyscalOn-TsyscalOff).
The plots are:
-
Tsys
(Deg K) for the alfa receiver band. Tsys in DegK is over plotted with
polA on the top frame and polB on the bottom frame (each pixel is plotted
in a different color). The cal
values are taken from the NFA measurements in the lab. The vertical
dashed lines are the edges of the 25 Mhz band passes. When scaling to kelvins,
the cal value at the center of each 25 Mhz band was used to scale the entire
25 Mhz band.
-
Tsys
(Deg K) by Pixel. Each pixel is plotted on a separate page. The top
frame is polA while the bottom frame is polB. The black line is Tsys in
degK. The red lines are Tsys/Tcal*10 to let you see how this ratio is changing
with freq (this plot does not depend on any measured cal values). The Green
line with * are the discrete cal values measured in the lab. The Blue lines
are the 3rd order fit to the discrete cal data (they are the values used
to convert to Kelvins).
The system temperatures range from 21K (probably a cal error) to 35 K.
The cal values were measured in the lab with the noise figure analyzer.
This is a first attempt at measuring the alfa cal values. We will redo
it using sky and absorber with alfa's LO chassis installed.
processing: x101/040422/calinp,calplot.pro
home_~phil